الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 120 |  |  |
| | | from | 120 | | |
Abstract
A drug-drug interaction (DDI) occurs when the effects of one drug are modified by the concomitant use of a second drug. Potential drug-drug interaction may be defined as the theoretical possibility of one drug physiologically altering the pharmacological effects of another drug, concomitantly prescribed.
The interaction between drugs may occur pharmacodynamically by synergistic reaction (i.e., 1+1 > 2) or antagonistic reaction (i.e. 1+1 <2) for the effect of the drug by another, or by the pharmacokinetic mechanism through modification of absorption, distribution, metabolism, and elimination of the drug from the site of action due to presence of another drug. Evaluation of potential DDIs is performed by investigating the clinical relevance of the interaction (severity, level of evidence, risk rating, and onset of action) and then by individual risk for potential DDIs on the patient according to his clinical situation.
Potential DDIs depend on many factors such as demographics like age and gender. Former studies reported polypharmacy as the main risk factor. It is known as “the prescription or consumption of two or more distinct medications for at least one day”. An American study conducted in 2017 concluded that 75.2% of children in pediatric ICU who receive 10 distinct drugs per day were exposed to at least one potential DDI. Longer pediatric ICU length of stay is associated with an increase in the prevalence of potential DDIs. Regarding medications, the risk of potential DDIs increases with some drug categories like antiepileptics, antibiotics, and endocrine drugs (mainly corticosteroids). Disease states and life-saving interventions like extracorporeal membrane oxygenation (ECMO) can alter organ function. ECMO adsorbs most of the drugs to the circuit resulting in sequestering them and altering their pharmacokinetics and consequently DDIs occurrence.
Early detection of potential DDIs is the golden solution to prevent and control DDIs. Healthcare workers should be aware of the complications of DDIs. Physicians should use the safest medication alternatives, apply therapeutic drug monitoring for medications with narrow therapeutic index, and periodical revising for the patient profiles and medication policies to change one of the interacting drug pairs to a non-interacting one, and apply time spacing of doses of the interacting pairs in addition to close monitoring of drug therapy.
Computerized potential DDI software is one of the tools that clinicians trust to review patients’ medication sheets. It could filter the clinically relevant interaction risks from insignificant ones and facilitate ignoring minor DDIs if not clinically relevant. The main drawback of these softwares is that till now they analyze the potential DDIs of the listed drugs by binary technique and ignore the net result of the impact of the drug combinations on the whole body. Several softwares are used for identifying potential DDIs in pediatric ICUs such as Lexicomp, Micromedex, fact and comparison, Medscape, and Epocrates. All mentioned softwares investigate potential DDIs in adults. Lexicomp software showed the best performance and highest scores of completeness (100%) and scope (97.0%) which was used in the current study.
This study aimed to study potential DDIs in children admitted to the intensive care units in pediatric hospitals in Alexandria. The specific objectives of the study were:
1. To assess potential DDIs and their level of severity among children admitted to ICUs in pediatric hospitals in Alexandria.
2. To identify the possible determinants of potential DDIs among children admitted to ICUs in pediatric hospitals in Alexandria.
3. To identify the main drug classes involved in severe DDIs.
A cross-sectional study was conducted in the ICUs of children’s hospitals affiliated to MoHP in Alexandria (El-Raml pediatric hospital and El-Anfoushy pediatric hospital). The target population was patients’ files attached prescriptions of infants and children aged between one month to ten years admitted to pediatric ICUs of both hospitals. Pretesting of the study tools was conducted in El-Raml hospital on 40 patients’ files to check possible obstacles during data collection. The time required for data collection from patients’ files and attached prescriptions for every child ranged from 15-25 minutes.
The minimum required sample was calculated using Epi Info 7.2. Based on the prevalence of 63.1% of potential DDIs in India, and 5.0% confidence limits, the minimum required sample size at a 95.0% confidence level was calculated to be 358 and rounded to 400 patients’ files and their attached prescriptions. The sample was proportionately allocated from the ICUs of the two hospitals according to the number of beds in each unit [to collect 144 patients’ files and attached prescriptions from El-Anfoushy hospital (9 beds and 17 rounding physicians) and 256 patients’ files and attached prescriptions from EL-Raml hospital (16 bed and 20 rounding physicians)]. Patients’ files were revised to collect medical prescriptions that contained two or more drugs excluding topical drugs. The researcher recruited the patients’ files consecutively until reaching the required sample size. All prescriptions in the patients’ files were considered because some patients may have more than one prescription during their stay in the pediatric ICU.
The data collection period was seven months which started on the beginning of July 2021 till the end of January 2022. The patients’ files and prescriptions were reviewed at the discharge of the patients to collect the following data: socio-demographic data, anthropometric measurement (weight, height), glomerular filtration rate (GFR) using the Counahan-Barratt Formula: 0.43* height (cm) /Serum Create(mg/dl)), present medical history and diagnosis, comorbid condition and premedication used (drug name and site of administration), laboratory abnormalities including renal, hepatic, and coagulation function, any present infection and its type from infection control records including periodical cultures on admission or during the hospitalization period. Also, regarding length of stay in ICU, case outcome at end of stay; (cure, referral (with the cause of referral), disability, or death). 24-hour prescription which included: prescribed drugs, number of drugs per prescription, and total number of drugs administered during the patient stay. Regarding drugs’ dosage forms, doses, and regimens, the number of prescribing physicians during patients’ stay in pediatric ICUs. Then, all drugs for each patient were entered, screened, and registered on Lexicomp software to check potential DDIs. Each monograph includes the following: significance rating (risk), interaction’s significance which included: onset, severity, and documentation, mechanism of interaction and all its conditions, management protocol, discussion, and references.
Data handling period took 5 months which started at the beginning of February 2022 till the end of July 2022. The collected data were revised for accuracy and completeness, cleaned, coded using Microsoft Excel, and analyzed using Statistical Package for Social Sciences (SPSS) version 24 software for tabulation and analysis. Testing for the normality of the data was done using One-sample Kolmogorov-Smirnov Test which revealed that most data did not follow normal distribution, and then the following statistical analysis was done:
• For all statistical procedures, the 5% level (p<0.05) was used as cut off value for statistical significance.
• Chi-squared tests were used to assess the associations between categorical variables in case of normal distribution.
• Fisher’s exact test was used when the Chi-square was invalid (20.0% or more of cells had an expected count less than 5).
• Mann Whitney test was used to test the significance of numerical variables in case of non-parametric distribution.
• Multivariable logistic regression analysis was done to determine the model shown by the data.
The study revealed the following main results:
The epidemiological characteristics of children admitted to pediatric ICUs in MoHP hospitals.
a. Demographics of children and their medical profiles
Regarding the epidemiological characteristics of children in the pediatric ICUs, age ranged from 1 month to 10 years (120 months) with a median of 5 (2-16.75). The majority of children (70.3%) were one year old or younger with 58.7% males, while toddlers (13-24 months) were less than one-fifth of the studied sample (12.7%) with 42.9% males. Early and middle childhood [(25-60 months) & (61-120 months)] constituted 10.5% and 6.5% of the sample respectively. Classification of age groups was derived from the American academy of pediatrics (AAP).
Regarding laboratory investigations, 56.2% of studied children had normal renal function, and 42.2% suffered from renal impairment, only 9.8 % of children had hepatic abnormalities. About 5.0 % of the studied sample had coagulation abnormalities. Almost two-thirds (68.3%) of children admitted to the pediatric ICUs had infections (such as sepsis, pneumonia, meningitis, and COVID-19) while 31.7% of them didn’t suffer from infections.
As regards children’s history before their admission to ICUs, almost one-fourth (101 cases) of children had been exposed to premedication before admission to the ICUs. Premedication was administered in 30.0% of those children in the hospital while 70.0% of them received their premedication at home. Regarding comorbid conditions, 38.5 % of children had comorbid conditions. Pneumonia was the most common diagnosis in the studied sample (34.3%) followed by bronchitis, Gastroenteritis (GE), and COVID-19 respectively. The median length of stay in the pediatric ICUs was 7 days (5-10). More than one-third of the children (36%) of children stayed for 5 days or less in the ICU unit while 44.0% of children stayed for 6-10 days. About two- thirds (61.5%) of children have been visited by 6-10 physicians during ICU stay with a median equal to 8 days (7-10.25).
b. Classification of medication classes used in pediatric ICUs.
The number of medications administered per day ranged from 2-14 medications per day with a median of 6 medications per day (5-8). Children who received 5-7 drugs per day constituted 51.0% of the studied sample. More than one-third (41.2%) of children have received 5-7 drugs, 32.5% have received 8-10 drugs, and 16.0% have received more than 10 drugs per total ICU stay with a median value equal to 7 medications per stay (6-9). The total number of drugs used was 136 drugs of different therapeutic categories. The most frequent therapeutic category was the anti-infectives, which constitute 35 medications, followed by cardiovascular drugs and GIT medications (18 and 16 medications respectively). All investigated cases received antibiotics; they received 2-6 antibiotics along their total stay. Most administered drugs were respiratory inhalators: ipratropium, budesonide, and salbutamol (albuterol) with frequency 6.2%, 5.8% and 5.6% respectively, while the second drug class used were antibiotics: ceftriaxone, cefotaxime, azithromycin, ampicillin-sulbactam and vancomycin with frequency ranged from 4.4% to 5.5%. The frequency of prescribed medications were 3043 medications 96.5% of them were accurate doses (tables 4.7, 4.8, and 4.9).
Prevalence of potential DDIs among medications prescribed to patients admitted to pediatric ICUs in MoHP hospitals and their classifications.
The prevalence of potential DDIs among children admitted to pediatric ICUs was 65.3%. The number of potential DDIs ranged from 1-7 drug interactions with a median value of 2(1-3). Almost two-thirds of children had been exposed to 1-2 potential DDIs. There was a total of 618 Potential DDIs cases for 160 DDI pairs that resulted from the coadministration of 136 medications (figure 4.3).
Concerning potential DDIs severity, 225 children were exposed to 442 potential DDIs for 129 moderate interaction pairs, while 130 children were exposed to 149 potential DDIs for 24 minor interaction pairs.
Regarding the level of documentation, an excellent level of documentation was found in 18 potential DDI pairs that detected 51times in 193 children while half of potential DDI pairs (50.0%) were of a fair level of documentation (80 DDI pairs) with frequency equal to 431 in 241 children.
According to the level of risk, the majority of potential DDIs were found in category C (monitor therapy). About three-quarters (73.1%) of DDI pairs (117 potential DDIs) had detected 287 times in 124 children. Category B (no action is needed) was expected 270 times in 173 children by 15.0% of DDIs (24 potential DDI pairs).
Regarding the onset of action, 145 (90.6%) potential DDIs on Lexicomp were found without documented data about the onset of action. These interactions constitute 96.1% of the total potential DDIs found (594 times) affecting 259 children (table 4.13).
The potential interactions of the highest frequency in children’s prescriptions resulted from the co-administration of azithromycin & salbutamol (14.2%), the co-administration of hydrocortisone with salbutamol (7.9%), and the co-administration of prednisolone with salbutamol (6.7%). The most predominant mechanism of action in the study was the pharmacodynamic mechanism. The clinical relevance of the potential DDIs was investigated for the first time in this study for every child according to his clinical profile (Table 4.16). The potential DDI (Azithromycin with Salbutamol) was clinically relevant in 59.1% of children who administered the combination despite its low clinical risk, while the combinations of dexamethasone with midazolam and amikacin with furosemide were clinically relevant for all children who received that combination (table 4.17).
Risk factors for potential DDIs among medications prescribed to patients admitted to pediatric ICUs in MoHP hospitals.
The study showed that potential DDIs were higher in children who were exposed to premeditations than in those who did not expose to them (79.2% vs 60.2%). This association was statistically significant (p=0.001). Concerning medications, children who administered 11-14 medications per day during their stay in the pediatric ICUs had the highest percentage of potential DDIs (95.7%). The number of children who had potential DDIs was directly proportional to the number of medications administered per day (87.0% in children who administered 8-10 medications per day, 65.7 % in children who administered 5-7 medications per day, and 24.7% in the children who administered 2-4 medications per day) with a statistically significant difference (p=0.0001).
Total medications administered during ICU stay were associated with potential DDIs. About 97.0% of children who had 11-14 drugs in ICUs had been exposed to potential DDIs. The proportion of children who were exposed to potential DDIs decreased by decreasing the number of medications during their stay in the pediatric ICUs (83.8% in the group of 8-10 drugs, 52.1% in the group of 5-7 drugs, and 9.8%in the group of 2-4 drugs per ICU stay). The association here was statistically significant (p=0.0001) (table 4.20). Regarding demographics, potential DDIs were associated with age and sex with statistically insignificant difference.
The Multivariate analysis model demonstrated that only two variables were significantly affecting the development of potential DDIs. These variables were diagnosis (particularly pneumonia) and the number of drugs per stay. Potential DDIs were 1.97 times more likely to occur among children who suffer from pneumonia than children with other diagnoses (OR=1.97) and 1.92 times more likely to occur with increasing drug exposure during ICU stay (OR= 1.92) (table 4.23).
Conclusion:
Based on the results of the present study, the following can be concluded:
- The prevalence of potential DDIs in pediatric ICUs was relatively high (65.3%).
- Most of the potential DDIs were of moderate severity, with a fair level of documentation, non-specific onset, and level B of risk.
- About 18.0% of prescription medications were inhalers (salbutamol, budesonide, and ipratropium), their share in the detected potential DDIs was approximately 35.0%. They were of minor severity.
- Polypharmacy and diagnosis of children admitted to pediatric ICUs (particularly pneumonia) are the variables that are most associated with the development of potential DDIs.
- Premedication was tested for the first time as a determinant for potential DDIs in pediatric population and it was statistically significant.
- Not all potential DDIs detected by Lexicomp were investigated in pediatric populations. Most of them were investigated in adults.
Recommendations:
A. Recommendations for the Ministry of Health and Population:
- Conduct surveillance to detect the occurrence of DDIs in children admitted to the pediatric ICUs and compare the results with the potential DDIs expected by the DDIs checkers softwares.
- Conduct training programs for healthcare workers about drug safety and DDIs complications and their management protocols.
- Broaden the role of clinical pharmacists to detect and manage polypharmacy in patients’ records.
- Provide more technical support for hospitals and supply healthcare workers with updated drug information sources for pediatric specialty and free DDI checkers softwares.
B. Recommendations for researchers
- Conduct studies to investigate the incidence rate of DDIs which were expected by the DDIs checkers in the Egyptian pediatric hospitals, and to evaluate the efficiency of the DDI checkers used.
- Conduct further studies to investigate the association between premedication and potential DDIs in children and to focus on the impact of medication type duration of administration before ICU admission and its site.
- Conduct studies to create DDI checker software that considers the administration of multiple medications instead of the current which considers the binary system.
- Conduct studies to investigate DDIs specifically in pediatric population and provide its results to the DDI check.